Physics-in-the-Loop: A Hybrid Agentic Architecture for Validated CAD Engineering Design

TL;DR

This paper introduces a hybrid AI architecture that integrates physical verification tools into the CAD design process, resulting in more complex and validated engineering designs.

Contribution

It proposes a novel agentic framework embedding knowledge-based engineering tools for physically validated CAD design, along with a new benchmark dataset and metrics.

Findings

Designs are 4.2 times more complex structurally

Compile rate improves by 3.5% over similar methods

System generates more physically verified CAD designs

Abstract

Large Language Models (LLMs) can generate Computer-Aided Design (CAD), yet lack physical comprehension required for reliable engineering design. Instead of attempting to implicitly learn physical laws from data, we propose a Hybrid Agentic-Physical Architecture that embeds validated knowledge-based engineering tools directly into the decision making loop of autonomous AI agents. In this framework, engineering design is formulated as a closed-loop, sequential decision making process guided by explicit physical verification. Based on a load case, dedicated agents iteratively plan, generate, evaluate, and revise engineering designs using knowledge-based tools as a feedback signal. We introduce a benchmark dataset and metrics for assessing functional validity in generative CAD. Our system generates more complex and physically verified designs, with a 4.2 increase in structural complexity and improving compile rate by 3.5% compared to similar agentic methods. The codebase, prompts and dataset will be made publicly available to support reproducibility and future research.